Shadow target data orientation analysis device and method

Shadow target data orientation analysis device and method

  • CN 111,444,752 A
  • Filed: 10/22/2019
  • Published: 07/24/2020
  • Est. Priority Date: 10/22/2019
  • Status: Active Grant
First Claim
Patent Images

1. A shadow target data orientation analysis device, the device comprising:

  • the shadow collecting device is connected with the data combining device and used for receiving the current combined image, taking pixel points of which the gray values are within a preset shadow gray distribution range in the current combined image as shadow pixel points, and removing isolated shadow pixel points in the current combined image based on the distribution positions of the shadow pixel points in the current combined image to obtain a plurality of residual shadow pixel points;

    the signal identification equipment is connected with the shadow acquisition equipment and is used for sending a shadow existence signal when the number of the plurality of residual shadow pixel points is greater than or equal to a preset number threshold;

    the signal identification equipment is further used for not sending out a shadow existence signal when the number of the plurality of the remaining shadow pixel points is smaller than the preset number threshold;

    the monitoring video equipment is arranged on the side surface of the shadowless lamp of the operating table and used for executing video recording action on a scene below the shadowless lamp so as to obtain and output a current video frame;

    the quantity identification equipment is connected with the monitoring video recording equipment and is used for receiving the current video recording frame and obtaining the quantity of pixel points of each line of the current video recording frame to be output as the image length;

    the instant segmentation equipment is connected with the quantity identification equipment and used for determining the size of a fragment area for image segmentation which is in direct proportion to the image length and segmenting the current video frame based on the determined fragment to obtain a plurality of sub-images;

    the GPU chip is connected with the instant segmentation equipment and used for receiving each subimage of the current video frame, sequencing the definitions of each subimage from small to large to obtain a definition sequence, taking the definition corresponding to a central sequence number in the definition sequence as a reference balance value, and performing definition improving processing on the subimage of an object in the current video frame by using the reference definition, wherein the higher the reference definition is, the fewer the times of the definition improving processing are performed;

    and the data combination equipment is connected with the GPU chip and is used for carrying out image combination on the sub-image without the object in the current video frame and the processed blocks obtained after the definition improvement processing is carried out on the sub-image with the object in the current video frame so as to obtain a current combined image corresponding to the current video frame, wherein one or more sub-images with the object in the current video frame are obtained, and one or more sub-images without the object in the current video frame are obtained.

View all claims
    ×
    ×

    Thank you for your feedback

    ×
    ×